A 34 Terabyte Partition—Yikes! A 1.7 Terabyte file—Ouch!

It wasn’t too long ago that a large IBM i server had one terabyte of DASD. Last week, a customer told me they had one partition with 34 terabytes of used DASD. Another customer told me that they have a physical file that is 1.7 terabytes (they’ve reached the maximum size allowed). These are just two examples of data growth on the IBM i platform. And, this trend will continue with more and more server consolidation. The good news is that, in both of these cases, Help/Systems can help.

• Robot/SPACE helps the first customer track their disk space—it highlights when a certain user, file, directory, or library is taking too much space in relation to the other data on the system. And, Robot/SPACE can automatically purge data that is old or not meaningful anymore.
• SEQUEL helps the second customer. They use the technology called file partitioning to divide their file into partitions and SEQUEL works with these partitions as if they were one file.

Over pizza the other day, Ron Siven (a long-time developer at Help/Systems) was explaining that he and his son were recently discussing large numbers. The conversation turned to how long it would take to count out loud to various numbers. Ron explained that, assuming it takes a second to say each number, a human could count to about 86,400 in approximately 24 hours. (Obviously, the counting gets harder as the number gets bigger.) Trust me on this—I don’t want anyone to see whether they can count to more than 86,400 in one day!

They determined that it would take a human 31 years (11,574 days) to count to 1 gigabyte (assuming you could stay awake for 31 years). So, how long would it take to count out loud to 34 terabytes? Here are the answers: A million (1 MB) seconds takes 12 days; a billion (1 GB) seconds takes 31 years; a trillion (1 TB) seconds takes 31,688 years. And, 34 TB takes 1,077,392 years.

The point is that we take a lot for granted as we store more and more data on our computer systems. This amount of data impacts the performance of queries, backups, and applications. Now, more than ever, we need to purge and clean up old data.

Have a great week.